The Advisor You Didn't Know You Had

I had an interview with Time Magazine a couple of weeks ago. I thought it went pretty well—you know, good questions, decent answers, felt conversational. Wrapped up feeling pretty good about it.

Then, on a whim, I had an idea: What if I asked AI what it thought of my performance?

I fed the transcript into my Claude Chief of Staff. What came back wasn't a pat on the back. It was a brutal, line-by-line takedown of everything I could have done better. The places I rambled. The moments I missed opportunities to drive my key points home. The questions I didn't fully answer. The examples I should have used but didn't. The follow up email I needed to send to make my points coherent.

It was honest in a way that almost no human would ever be with me.

When Humans Are Too Nice

The challenge of seeking feedback from other people is, they're kind. Even the smartest, most well-intentioned people in your life will pull their punches. They'll say, "You did a good job," because they don't want to hurt your feelings or damage the relationship. And honestly? You probably do the same thing for them. (I know I do.)

Think about it. When's the last time someone gave you truly brutal, objective feedback on your work? Not criticism with an agenda. Not vague suggestions wrapped in a compliment sandwich. But the kind of unvarnished assessment that actually helps you get better?

For most of us, the answer is never. Or maybe once, from that one person who didn't care about making you uncomfortable—and you probably didn't love hearing it at the time. (But looking back, they were probably right.)

A generation ago, if you wanted access to that kind of honest, expert analysis, you had to be independently wealthy. You needed a personal PR advisor. A communications coach on retainer. Someone whose literal job was to watch your interviews, dissect your performance, and tell you the truth.

Who am I to have that kind of advisor? (I mean, I'm an adjunct professor who writes a weekly AI newsletter. Not exactly in the "personal PR team" demographic.)

Turns out, now I do. And so do you.

An Emerging Class of Possibilities

What I realized after that Time interview is that there's this emerging class of activities that used to require being independently wealthy—and now they're just... available. To everyone. For the cost of a subscription. (Twenty bucks a month. That's less than most people spend on coffee.)

Personal PR analyst? Check. Executive coach? Available 24/7. Strategic advisor who knows everything about your work and can spot opportunities you're missing? That's just Claude or ChatGPT, if you know how to work with it.

The week before the Time interview, I'd had a conversation with a heavy hitter—someone who sits on multiple Fortune 50 boards. Ordinarily, a conversation like that would just be... a conversation. I'd reflect on it while driving home, maybe jot down a few notes, and move on. That's what I've always done.

But I was on an escalator at the airport (of all places), and I had a thought: I bet AI could analyze the conversation and give me feedback.

I pulled out my phone, uploaded the transcript, and asked for an honest assessment. What came back was eye-opening. AI didn't just tell me what went well (though it did). It gave me a brutally accurate read on missed opportunities, places where I could have been clearer, and—here's the kicker—a strategy for following up that I never would have employed on my own.

I would have lacked the bandwidth to even sufficiently reflect on the conversation, let alone assess it accurately, let alone scope out a way to follow up based on the actual conversation.

That experience led directly to my idea with the Time interview. Using AI to improve one conversation sparked the insight to use AI to improve a completely different kind of conversation. That's the first level of what I call "use AI to use AI."

The Meta AI Loop

There are actually two levels to this "use AI to use AI" idea, and both matter. (And yes, I know the phrase "use AI to use AI" sounds a bit like those recursion jokes programmers love. Bear with me.)

Level 1: Your AI use sparks new AI uses.

As you work with AI, you start seeing possibilities you didn't see before. I used AI to analyze a board conversation, which made me think, "Wait, could I do this for media interviews too?" One use case reveals another. Your imagination expands as you actually engage with the technology.

Though the idea is deeply backed by cognitive science, the suggestion isn't theoretical. It's what actually happened to me. And it's what happens to everyone who moves beyond treating AI like a tool and starts treating it like a thinking partner. (Which, by the way, is the shift that matters most.)

Level 2: AI can teach you how to use AI.

Even wilder? If you're stuck on how to use AI for something, you can ask AI itself to help you figure it out.

Let's say you want to analyze your meeting performance, but you don't know how to frame the request. You could spend time Googling "best prompts for meeting analysis" and reading articles about prompt engineering. Or you could just ask:

"You're an AI expert. I have transcripts from my recent meetings and I want to use AI to analyze my contributions and get better. How should I approach this? What should I ask for? What context do you need?"

The AI will interview you, figure out what you're trying to accomplish, and help you craft the exact prompt you need. It's the first technology in history that can teach you how to use itself better.

I've seen this play out over and over. Someone in one of my workshops will say, "I want to do X with AI, but I don't know how." And I'll say, "Have you asked ChatGPT how to do it?" The lightbulb moment is instant. They literally never thought to ask.

The Cold War Judge Prompt

Here's my favorite phrase for getting honest feedback: "Act like a Cold War era Russian Olympic judge."

I know, I know. It's ridiculous. (And to be clear, this has nothing to do with actual Russians—it's just a funny cultural reference that provokes a specific response.) But it works. There's something about that image—the unsmiling face, the clipboard, the complete unwillingness to be impressed—that signals to the AI exactly what you want. You're not looking for encouragement. You're looking for the truth.

Most people use AI to hear they did a good job. They want the validation. They want the rubber stamp of approval. And AI will absolutely give you that if that's what you ask for. (AI is very accommodating that way.)

But if you want to get better? If you actually want to improve? You need to tell the AI that's what you want. You need to raise your expectations and be explicit about the kind of feedback you're seeking.

Try this prompt next time:

"I'm going to share a transcript of [my presentation/interview/meeting]. Act like a Cold War era Russian Olympic judge—be exacting and brutally honest with me. I want to improve, not hear that I did well. Analyze my performance and tell me specifically where I fell short and how I could have been better. Don't pull any punches."

You'll be amazed at what comes back.

Try This Right Now

Here's something you can do in the next five minutes that will blow your mind. (I'm serious. Stop reading and actually do this.)

If you use Microsoft Copilot (or any AI that passively records your meetings), open it up and copy/paste this prompt:

"You're a professional communications coach and executive advisor. Please review the last five meetings I've participated in. Analyze my contributions to those meetings specifically—not the meetings overall, but how I showed up. Then give me three specific recommendations for how I can be a better teammate, communicator, or contributor."

The insights you get will be far more valuable than what any human in your organization could give you. Not because the AI is smarter than your colleagues (it's not), but because it doesn't have to worry about your feelings, your ego, or the political dynamics of your workplace.

It can just tell you the truth.

And truth is what you need if you actually want to improve.

What Feedback Are You Missing?

Serious question: What honest feedback are you not getting right now because the humans in your life are too polite?

Where are you walking away from important moments—interviews, presentations, difficult conversations, pitches—thinking you did fine, when the reality is you missed opportunities you didn't even see?

Who among us doesn't have blind spots? Who among us couldn't benefit from the kind of brutally honest advisor that used to be reserved for people with personal coaches on retainer?

The advisor exists now. You already have access. The only question is whether you're willing to hear what it has to say.

(Real talk? If you're not willing to hear it, that's probably the surest sign you need to.)

Related: Have You Tried AI? The Answer Every Leader Should Give
Related: Don't Keep the AI Expert Waiting
Related: Use AI to Use AI
Related: What’s Your Question: The Missing Skill That’s Killing Innovation

Join over 28,147 creators & leaders who read Methods of the Masters each week.

Next
Next

The AI Multiplier: Why Your "Organic Capabilities" Matter More Than Ever